J. COLLINA N D F. P. LOSSING
20G4
narily docs not dissociate in liquid phase. I t is to he noted t h a t the only saturated hydrocarbons resulting from reactions 2 or 3 are branched and, in some cases, cyclic. This is in accord with the results : the ozone-resistant dimer contained no ndodecane. The yield of hydrogen, which is not balanced b y the extent of unsaturation i n the dimers, trimers and tetraiiiers necessitates the assumption t h a t the Iiiglier 1mlyliiers must contain rings or inore than CJnC dou1)lc 11ontl per Inolecule. -1 reaction scheme involving scission of n car~,r)n.-hyclrr)genboric1 in an olefin yielding a hg-drogen atom and a radical would lie inconipatiblc with the experimental findings. IIon-cver, the production of hydrogen in the rndiolysis of paraffins has been attributed largely to a iiioleculnr p r o c e ~ s . ~ ' ?Molecular ~,~~ pmduction of (2.1) I-. \I. 1 ) i i r f m n n . J . P l i y s . C h e m . , 6 0 , 8?G (l'J.jf;), ( 2 i ) R IT S r l i r i l ~ r ibi , / . . 6 0 , 3 5 1 (1951;).
VOl. 81
hydrogen from an olefin molecule would result in the forination of a diene which would be very reactive in chain polymerization. Acknowledgments.---U'e wish to acknowledge that this work was carried out at the suggestion and with tlie encouragement of the late Professor AT. S . Kharasch. Encouragement from Professor IV.G. Brown after Professor Kharasch's death is gratefully acknowledged. We are indebted to Drs. D. P.Stevenson and D. 0. Schissler of Shell Developiiient Company for the interpretation of the inass spectra and to Dr. Stevenson for molecular orbital calculations that apply here. Discussions with I h . G. 11.Coppinger, E. R. Bell and T. P. Iiudy regarding the work were also very helpful. C i r ~ c ~ aII.I.ISOIS o, ~ ' M E R V V I I , I . R , CAL.
Ionization Potentials of Some Olefins, Di-olefins and Branched Paraffins BY J. COLLIS~ X X D F.P.LOSSIXG R E C E I V E0DC . l ) J . J . l l i t c h e l l a n d 1'. r . C o l e m a n , t b I , i . , 17, 41 (1919). /I?) R E Fox a n d A . L a n g e r . ibrd , 18, $60 (1950). ( 7 ) I). P Stevenson a n d J . A . Htp,lle, Tnrs J O I . K N A L , 64, 1588 ( I (1 121, is) R . i'.B a k e r arid J . T . ' r a t e , P h y s . R e v , 63,51-1 (1938). ( 9 ) T h e s e preliminary d a t a were m a d e ni,:rilable t o t h e aiithor t l i r u u g l i t h e courtesy of nr. E;. Wat.inahe. ( l o ) J. L. Franklin, J , Chem. P h y s , , 22, 1301 (1954).
May 5, 1050
IONIZATION POTENTIALS OF OLEFIXS,Dr-or2Errss I\SD BRXSCIIED PAR.IFFISS TAIILE I1
TABLE I IOSIZATIOSPOTESTIALS OF R R A S C I I E D PARAFFISS --Electron Compound
Isobutane
2-Methylbutane 2-hlethylpentane 3-Methylpentane 2,3-Di-mcthylbutane
Lit.
impactThis work
2065
Photoionizntion9
--Calcd.lo-
10.3 =I= 0.24 11.005 10.7 f 0.2$ 10.34 f 0 . 1' 9 . 8 f 0 29
l n . i 9 10.55 10.82,10.03
IO. 1 f 0 . 2 '
10.60 lCJ.30 10.67, 10.-13
10.0 f 0.Z4
1 0 . 3 1 10.09
. , ... .. . .
9.8 f0.24
10.30 10.00
,
1 0 . 1 zk 0 . R 4
10.2-4 10.00
IOYIZATIOS POTEYTIALS OF OLEFISS Compound
Electron i m p a c t This Lit. work
Photoioni7.ation
Spectroscopy
Calcd.
9.7012
9.281"
9 . 281° . . . . , . ..
10.48,10.21
ment with the value of Fox and Langere6 T h e values of Mitchell and Coleman: and of Baker and Tate8 appear to lie outside the expected range. For 2-methylpentane and 3-inethylpentane the ionization potentials are found to be nearly the same, and appreciably lower than for 2-methylbutane, as would be expected from structural considerations. T h e photoionization data show a siinilar relation. T h e ionization potential of 2 , s dimethylbutane is lower than that of the other branched hexanes, rather than higher as indicated by llitchell and Coleiiian's data. This is i n agreement with the differencc shown by the photoionization results and is again the relation which ~vould be predicted. In view of the difference b e t w e n the electron impact and the photoionization data, a choice between the two sets of calculated values cannot be made, b u t it may be noted that the magnitude of the decrease in ionization potential with increasing substitution is reproduced very closely in both sets. (b) Olefins.-The ionization potentials found for a number of straight and branched olefins are given in Table 11. For comparison, the electron impact values published earlier,5,6e11-14 values obtained by p h o t o i o n i ~ a t i o n , ~and , ' ~ by ultraviolet spectroscopy,16and values calculated by molecular orbital method^^^,^^ also are shown. il comparison of the present measurements with the photoionization data shows, as before, a close parallelism. For the olefins the difference between values obtained by the two methods, 0.1-0.2 v., is smaller than for the paraffins. The agreement among the electron impact data for 1-butene is satisfactory. For isobutene, the value found in this work is in good agreement with t h a t found by Honig'J and with the photoionization r e ~ u l t . ~This supports (11) D. P. Stevenson, THISJ o u n s a ~ 65, , 209 (1913). (12) V. H . Dibeler, J . Research X o f l . Birr. S f a > l d a r d s , 38, 329 (1947). (13) R. E. Honig, J . Chsm. P h y s . , 16, 10.5 (1948). (14) D. P. Stevenson and J. A. Hipple, Tms J O U R N A L , 64, 2769 (19 $ 2 ) . (15) K. W a t a n a b e , J. Chem. P h y s . , 26, .5$2 (1957). (16) W . C. Price and W. T. T u t t e , Pvoc. R o y . Soc. ( L o l i d o n ) ,A174, 207 (1940). (17) G . G . Hall, Trans. Faraday SOC.,49, 112 (1953).
9.27)' 9 . 28Io 9 . 2917 9.071Q ...
...
... ... S.8l10 S . 85'; 8 ,4O10 8.52l7
the prctlietioii from structural considerations that the ionizxtion potential of isohutcne would he very close to those of cis- and fvaus-2-butene. The evidence therefore favors a vertical ionization potentid uf 9.2-9.3 v. for i s o h t e n c , and the lower \-dues of about S . 7 v.6,11 nppew to lie outside the espectcd rnnge. For ci.7- :i~idfrans-?-butene the electron impact, ~)Iiotoioiiizatioiiant1 calculated values are in quite close ayrceinent. The electron impact values suggest that I ( t r a ~ i s )is slightly less than I (cis), but this is not confirmed by the photoionization data. The value of 9.2 v. for I ('-butene) obtained from ultraviolet spectroscopy by Price and Tutte16 some years ago is in escellent agreement with the other data, as are the values for 3-methyl-2-butene and tetrainethylethylene. One particularly striking feature of the olefin data is the success of the calculations of Hall" and of Franklin'O in reproducing the measured ionization potentials to within 0.1 v. Xn interesting consequence of structural effects can be seen by comparing I (2-butene) with I (1butene), the foriner being about 0.5 v. less. Similarly I (2-pentene) is about 0.6 v. less than (1pentene). X comparable decrease (0.5v.) occurs on going from 1-butene to isobutene and from 1pentene to 2-methyl-1-butene (0.47 v.). These changes are consistent with an increase in hyperconjugational effects as methyl groups are added around the double bond. (c) Cycloolefins.-The ionization potential data for cycloblefins are summarized in Table 111. The ionization potential of the interesting compound cyclopropene has not been measured previously. T h e observed value, 9.95 v., is nearly the same as the ionization potential of propylene, 9.84
J. COLLINAND F. P. LOSSING
2066
v.la It may be noted that the ionization potential of cyclopentene found in this work, 9.27 v., is similarly close t o the ionization potential of 2pentene, 9.11 v., and that of cyclohexene, 9.18 v., is close to t h a t of 2-hexene1 9.16 v.I3 and 3-hexene, 9.12 v. l 3 The agreement of the value reported here
available values from the literaturea~16~21-26 have been included. As would be expected, the only conjugated pentadiene has an ionization potential TABLE IV IONIZATION POTENTIALS OF DI-OLEFINS
TABLE 111 IONIZATION POTENTIALS OF CYCLOOLEFINS Electron impact This Compound
Lit.
work
. . . . 9.95 Cyclopropene Cyclopentene 10. 218 9 . 2 7 Cyclohexene 9 . 718 9 . 1 8 9 . 24IB
Photoionization
.... 9 .O l g
8.94516
Compound Allene
Spectroscopy
Calcd.
.. ..
...
9 . 2j6
9 . 161°
...
for cyclopentene with the photoionization value, 9.01 v . , ~suggests that the earlier electron impact result of Hissells is considerably too high. For cyclohexene the agreement with the electron impact value of Morrison and Nicholson,1gthe photoionization value, l5 the spectroscopic value16 and the calculated valuelo is quite satisfactory. (d) Di-olefins.--The ionization potentials found for four pentadienes and for 1,5-hexadiene are given in Table IV. For comparison, values for allene and the butadienes measured previouslyz0 and the (18) J . Hissel, Bull. soc. roy. sci. Lihge, 21, 457 (1952). (19) J. D. Morrison a n d A . J. C. Nicholson, J . Chem. Phys., 20, 1021 (1952). (20) J . Collin a n d F. P. L o s i n g , THISJ O U R N A L , 79, 5848 (1957).
Vol. 81
1,z-Butadiene 1.5-Butadiene
1.2-Pentadiene 2 , 3 - Pen tadiene 1,3-Pentadiene 1.4-Pentadiene 1.6-Hexadiene
Electron impact This Lit. work 10.1620 10.02' 9 5720 9.1820 9 28 9.211s
,.
,
,
94?
....
9.26 8.68 9.5s 9.51
.... ....
SpectrosCOPY
Calcd.
10.192'
..
....
,._.
Photoionization
9 07"
cis8.752' t r a n s 9.072'
........
........ .......
........
........
... ... ...
........
...
.......
...
t r a n s 8 8126
........
9 2810
much lower (0.5 v.) than the three non-conjugated ones. The differences among the latter are small and probably result from hyperconjugational effects. OTTAWA, CANADA (21) J . Delfosse a n d W.Bleakney, Phrs. Rev., 56, 256 (1939). ( 2 2 ) L. H . Sutcliffe and A. D. Walsh, J . Chem. Soc., 899 (1952). ( 2 3 ) T. M , Sugden and A. D. Walsh, Trans, Faraday Soc., 41, 76 (1 945). (24) W, C. Price and 4 . D. Walsh, Proc. Roy, Soc. (London), 8174, 220 (1940). ( 2 5 ) N. R . Hush and J . 4 . Pople, Trnns. Faraday Soc., 51, 600 (1955).
